On the Premise of a Swarm Guidance Ontology for Human-Swarm Teaming

Abstract

Effective Human-Swarm Teaming (HST) relies on bi-directional information flow between the human and the swarm. Systems with human control or oversight rely on information flow from the swarm to the humans to inform decisions, while information that flows back from humans is only that necessary for actuation, which remains primarily physical. To unlock the full potential of HSTs, the augmentation must extend into the overall logic of teaming, including both the human’s and machine’s cognitive domains, whereby an AI-equipped robot teammate is capable of complex cognitive functions. The effectiveness of HST will need a sufficient level of transparency in the interaction space formed by the bi-directional information flow between the human and the swarm. This transparency must continuously and constructively interpret the information exchanged between the human and the swarm to afford both cognitive agents with the capacity to form shared understanding and situation awareness, and thus, facilitating effective teaming through trust. An ontology is one formal representational construct that enables bi-directional interpretation, thus, transparency. In this paper, we conceptualise and present a meta-ontology for transparent HST interactions.

Onto4MAT: A Swarm Shepherding Ontology for Generalized Multiagent Teaming

“…Throughout the first two stages in the design method, we identified a set of candidate terms and definitions that could be integrated from existing metaontologies. Of particular note, the Human-Swarm Teaming Guidance Ontology (HST-GO) [39], Swarm Ontology for Shepherding [12], and SOSA: a lightweight ontology for sensors, observations, samples, and actuators [40] formed the core of existing ontologies that provided definitions of terms whose semantics and implementation aligns with the terms identified in our conceptualisation of the domain. Ontology implementation is the codification of the ontology using formal representations in Prolog, Ontolingua, C++, or XML [38], or more recently, Python [41].…”

“…• Academic and domain literature containing the foundational concepts of shepherding and teaming systems. These covered shepherding from the modelling perspectives, see for instance [12], [34], [39], domain application perspectives through work such as Williams (2007) [58], and through to frameworks for HST, for example [11]. • Domain expert exposure through site visits to agricultural settings where biological shepherding occurs.…”

Swarm analytics: Designing information markers to characterise swarm systems in shepherding contexts

“…The second challenge motivated a line of research on activity recognition of human-swarm interaction (Hepworth, 2021), as well as designing ontologies to represent the space of concepts lying between humans and the swarm (Abbass & Hunjeta, 2021a; Baxter et al, 2021; Hepworth et al, 2022), contributing to a holistic theory to inform how humans and swarm should interact (Hasbach & Bennewitz, 2021). The third challenge motivated the design of contextual indicators that could be extracted from the sensorial information to guide the swarm.…”

“…The proposed swarm markers offer three extra advantages. The first is through the lens of the swarm agents, enabling activity recognition of other agents and the collective (Baxter et al, 2021). The second is through the lens of an external observer who can classify behaviours and infer intents (Hepworth, 2021).…”